CS 356: Computer Network Architectures Lecture 22: Internet Quality - - PowerPoint PPT Presentation

CS 356: Computer Network Architectures Lecture 22: Internet Quality of Service [PD] Chapter 6.5

Xiaowei Yang xwy@cs.duke.edu

2

Overview

• Network Resource Allocation
• Congestion Avoidance
• Why QoS?

– Architectural considerations

• Approaches to QoS

– Fine-grained: Integrated services

• RSVP

– Coarse-grained:

• Differentiated services
• Next lecture

Internet Quality of Service

4

Motivation

• Internet currently provides one single class of

“best-effort” service

– No assurance about delivery

• Many existing applications are elastic

– Tolerate delays and losses – Can adapt to congestion

• “Real-time” applications may be inelastic

5

Inelastic Applications

• Continuous media applications

– Lower and upper limit on acceptable performance – Below which video and audio are not intelligible – Internet telephones, teleconferencing with high delay (200 - 300ms) impair human interactions

• Hard real-time applications

– Require hard limits on performance – E.g., industrial control applications

• Internet surgery

6

Design question #1: Why a New Service Model?

• What is the basic objective of network design?

– Maximize total bandwidth? Minimize latency? Maximize ISP’s revenues? – the designer’s choice: Maximize social welfare: the total utility given to users (why not profit?)

• What does utility vs. bandwidth look like?

– Must be non-decreasing function – Shape depends on application

7

Utility Curve Shapes

• Stay to the right and you

are fine for all curves

BW U Elastic BW U Hard real-time BW U Delay-adaptive

8

Playback Applications

• Sample signal à packetize à transmit à buffer à playback

– Fits most multimedia applications

• Performance concern:

– Jitter: variation in end-to-end delay

• Delay = fixed + variable = (propagation + packetization) + queuing
• Solution:

– Playback point – delay introduced by buffer to hide network jitter

Characteristics of Playback Applications

• In general lower delay is preferable
• Doesn’t matter when packet arrives as long as

it is before playback point

• Network guarantees (e.g., bound on jitter)

would make it easier to set playback point

• Applications can tolerate some loss

10

11

Applications Variations

• Rigid and adaptive applications

– Delay adaptive

• Rigid: set fixed playback point
• Adaptive: adapt playback point

– E.g. Shortening silence for voice applications

– Rate adaptive

• Loss tolerant and intolerant applications
• Four combinations

13

Applications Variations

Really only two classes of applications 1) Intolerant and rigid 2) Tolerant and adaptive Other combinations make little sense 3) Intolerant and adaptive

• Cannot adapt without interruption

4) Tolerant and rigid

• Missed opportunity to improve delay

Design question 2: How to maximize V = ∑ U(si)

• Choice #1: add more pipes
• Choice #2: fix the bandwidth but offer

different services

– Q: can differentiated services improve V?

15

If all users’ utility functions are elastic

• ∑ si = B
• Max ∑ U(si)

Bandwidth U

Does equal allocation of bandwidth maximize total utility?

Elastic

16

Design question: is Admission Control needed?

• If U(bandwidth) is concave

à elastic applications

– Incremental utility is decreasing with increasing bandwidth

• U(x) = log(xp)
• V = nlog(B/n) p= logBpn1-p

– Is always advantageous to have more flows with lower bandwidth

• No need of admission control;

This is why the Internet works! And fairness makes sense BW U Elastic

17

Utility Curves – Inelastic traffic

BW U Hard real-time BW U Delay-adaptive

Does equal allocation of bandwidth maximize total utility?

18

Is Admission Control needed?

• If U is convex à inelastic

applications

– U(number of flows) is no longer monotonically increasing – Need admission control to maximize total utility

• Admission control à deciding

when the addition of new people would result in reduction of utility

– Basically avoids overload BW U Delay-adaptive

Incentives

• Who should be given what service?

– Users have incentives to cheat – Pricing seems to be a reasonable choice – But usage-based charging may not be well received by users

Over provisioning

• Pros: simple
• Cons

– Not cost effective – Bursty traffic leads to a high peak/average ratio

• E.g., normal users versus leading edge users

– It might be easier to block heavy users

Comments

• End-to-end QoS has not happened
• Why?
• Can you think of any mechanism to make it

happen?

22

Approaches to QoS

• Fine-grained:

– Integrated services

• RSVP
• Coarse-grained:

– Differentiated services

23

Components of Integrated Services

• 1. Service classes

What does the network promise?

• 2. Service interface

How does the application describe what it wants?

• 3. Establishing the guarantee

How is the promise communicated to/from the network How is admission of new applications controlled?

• 4. Packet scheduling

How does the network meet promises?

24

• 1. Service classes

What kind of promises/services should network

• ffer?

Depends on the characteristics of the applications that will use the network ….

25

Service classes

• Guaranteed service

– For intolerant and rigid applications – Fixed guarantee, network meets commitment as long as clients send at match traffic agreement

• Controlled load service

– For tolerant and adaptive applications – Emulate lightly loaded networks

• Datagram/best effort service

– Networks do not introduce loss or delay unnecessarily

26

Components of Integrated Services

• 1. Type of commitment

What does the network promise?

• 2. Service interface

How does the application describe what it wants?

• 3. Establishing the guarantee

How is the promise communicated to/from the network How is admission of new applications controlled?

• 4. Packet scheduling

How does the network meet promises?

Service interfaces

• Flowspecs

– TSpec: a flow’s traffic characteristics

• Difficult: bandwidth varies

– RSpec: the service requested from the network

• Service dependent

– E.g. controlled load

28

A Token Bucket Filter

Operation: – If bucket fills, tokens are discarded – Sending a packet of size P uses P tokens – If bucket has P tokens, packet sent at max rate, else must wait for tokens to accumulate

Tokens enter bucket at rate r Bucket depth b: capacity of bucket

29

Token Bucket Operations

Tokens Packet Overflow Tokens Tokens Packet

Enough tokens à packet goes through, tokens removed Not enough tokens à wait for tokens to accumulate

30

Token Bucket Characteristics

• In the long run, rate is limited to r
• In the short run, a burst of size b can be sent
• Amount of traffic entering at interval T is

bounded by:

– Traffic = b + r*T

• Information useful to admission algorithm

31

Token Bucket Specs

BW Time 1 2 1 2 3 Flow A Flow B

Flow A: r = 1 MBps, B=1 byte Flow B: r = 1 MBps, B=1MB

TSpec

• TokenBucketRate
• TokenBucketSize
• PeakRate
• MinimumPolicedUnit
• MaximumPacketSize

33

Service Interfaces: RSpec

• Guaranteed Traffic

– TokenRate and DelayVariation – Or DelayVariation and Latency

• Controlled load

– Type of service

34

Components of Integrated Services

• 1. Type of commitment

What does the network promise?

• 2. Service interface

How does the application describe what it wants?

• 3. Establishing the guarantee

How is the promise communicated to/from the network How is admission of new applications controlled?

• 4. Packet scheduling

How does the network meet promises?

35

RSVP Goals

• Used on connectionless networks

– Robust – Should not replicate routing functionality – Should co-exist with route changes

• Support for multicast
• Modular design – should be generic “signaling”

protocol

• Approaches

– Receiver-oriented – Soft-state

36

RSVP Service Model

• Make reservations for simplex data streams
• Receiver decides whether to make reservation
• Control msgs in IP datagrams (proto #46)
• PATH/RESV sent periodically to refresh soft state

37

PATH Messages

• PATH messages carry sender’s Tspec

– Token bucket parameters

• Routers note the direction PATH messages arrived

and set up reverse path to sender

• Receivers send RESV messages that follow reverse

path and setup reservations

• If reservation cannot be made, user gets an error

38

RESV Messages

• Forwarded via reverse path of PATH
• A receiver sends RESV messages

– TSpec from the sender – Rspec

Admission control

• Router performs admission control and

reserves resources – If request rejected, send error message to receiver – Guaranteed service: a yes/no based on available bandwidth – Controlled load: heuristics

• If delay has not exceeded the bound last

time after admitting a similar flow, let it in

40

Soft State to Adapt to Routing Changes

• Problems: Routing protocol makes routing

changes

• Solution:

– PATH and RESV messages sent periodically – Non-refreshed state times out automatically

• Ex: a link fails. How is a new reservation

established?

Merging multicast reservations

A requests a delay < 100ms B requests a delay < 200ms

42

Components of Integrated Services

• 1. Type of commitment

What does the network promise?

• 2. Service interface

How does the application describe what it wants?

• 3. Establishing the guarantee

How is the promise communicated to/from the network How is admission of new applications controlled?

• 4. Packet scheduling

How does the network meet promises?

Packet classification and scheduling

• 1. Map a packet to a service class

– (src addr, dst addr, proto, src port, dst port)

• 2. Use scheduling algorithms to provide the

service

– An implementation issue

44

Scheduling for Guaranteed Traffic

• Use WFQ at the routers

– Q: will DRR work?

• Each flow is assigned to its individual queue
• Parekh’s bound for worst case queuing delay = b/r

– b = bucket depth – r = rate of arrival

45

Controlled Load Service

Goals:

• Isolation

– Isolates well-behaved from misbehaving sources

• Sharing

– Mixing of different sources in a way beneficial to all

Possible Mechanisms:

• WFQ

– Aggregate multiple flows into one WFQ

46

Unified Scheduling

• Scheduling: use WFQ in routers

Controlled Load Class I Controlled Load Class II Best Effort Guaranteed Service Guaranteed Service

Scalability

• A lot of requests and state!
• ISPs feel it is not the right service model for them!
• Per-flow reservation/queue

– OC-48 link 2.5Gbps – 64Kbps audio stream – à 39,000 flows – Reservation and state needs to be stored in memory, and refreshed periodically – Classify, police, nd queue each flows

Comments on RSVP

• Not widely deployed as a commercial service
• Used for other purposes

– Setting up MPLS tunnels etc.

49

Summary

• Why QOS?

– Architectural considerations

• Approaches to QoS

– Fine-grained: Integrated services

• RSVP

– Coarse-grained:

• Differentiated services
• Next lecture:

– DiffServ – Net Neutrality

DiffServ

51

Motivation of DiffServ

• Analogy:

– Airline service, first class, coach, various restrictions on coach as a function of payment

• Economics and assurances

– Pay more, and get better service – Best-effort expected to make up bulk of traffic, – Revenue from first class important to economic base – Not motivated by real-time or maximizing social welfare

52

Basic Architecture

• Agreements/service provided within a domain

– Service Level Agreement (SLA) with ISP

• Edge routers do traffic conditioning

– Shaping, Policing, and Marking

• Core routers

– Process packets based on packet marking and defined per hop behavior (PHB)

• More scalable than IntServ

– No per flow state or signaling

DiffServ Architecture Example

AT&T

UNC

Duke

Shaping, policing, marking Per-hop behavior

54

Per-hop Behaviors (PHBs)

• Define behavior of individual routers rather than end-

to-end services; there may be many more services than behaviors

– No end-to-end guarantee

• Multiple behaviors – need more than one bit in the

header

• Six bits from IP TOS field are taken for Diffserv code

points (DSCP)

55

Per-hop Behaviors (PHBs)

• Two PHBs defined so far
• Expedited forwarding aka premium service (type P)

– Possible service: providing a virtual wire

• Assured forwarding (type A)

– Possible service: strong assurance for traffic within profile and allow source to exceed profile

56

Expedited Forwarding PHB

• Goal: EF packets are forwarded with minimal delay and loss
• Mechanisms:

– User sends within profile and network commits to delivery with requested profile – Rate limiting of EF packets at edges only, using token bucket to shape transmission – Priority or Weighted Fair Queuing

57

Assured Forwarding PHB

• Goal: good services for in-profile traffic
• Mechanisms:

– User and network agree to some traffic profile

• How to define profiles is an open/policy issue

– Edges mark packets up to allowed rate as “in- profile” or low drop precedence – Other packets are marked with one of two higher drop precedence values – Random Early Detection in/out queues

DiffServ Architecture Example

AT&T

UNC

Duke

Shaping, policing, marking Per-hop behavior Edge Core

59

Edge Router Input Functionality

Packet classifier Traffic Conditioner 1 Traffic Conditioner N Forwarding engine

Arriving packet

Best effort

Flow 1

Classify packets based on packet header

60

Traffic Conditioning

Wait for token

Set EF bit

Packet input Packet

• utput

Test if token

Set AF “in” bit

token No token

Packet input Packet

• utput

Drop on overflow

Router Output Processing

• Two queues: EF packets on higher priority queue
• Lower priority queue implements RED “In or

Out” scheme (RIO)

61

What DSCP? If “in” set incr in_cnt High-priority Q Low-priority Q If “in” set decr in_cnt RIO queue management

Packets out EF AF

Router Output Processing

• Two queues: EF packets on higher priority queue
• Lower priority queue implements RED “In or

Out” scheme (RIO)

62

What DSCP? If “in” set incr in_cnt High-priority Q Low-priority Q If “in” set decr in_cnt RIO queue management

Packets out EF AF

63

Red with In or Out (RIO)

• Similar to RED, but with two separate probability

curves

• Has two classes, “In” and “Out” (of profile)
• “Out” class has lower Minthresh, so packets are

dropped from this class first

– Based on queue length of all packets

• As avg queue length increases, “in” packets are also

dropped

– Based on queue length of only “in” packets

64

RIO Drop Probabilities

65

Pre-marking and traffic conditioning

first hop router internal router edge router CEO edge router

ISP Company A

Unmarked packet flow Packets in premium flows have bit set Premium packet flow restricted to R bytes/sec Policing

66

Edge Router Policing

Arriving packet

Is packet marked? Token available? Token available? Clear “in” bit Drop packet

Forwarding engine AF “in” set EF set

Not marked no no

Remarks on QoS

• “Dead” at the Internet scale
• Areas of success

– Enterprise networks – Residential uplinks – Datacenter networks

Conclusion

• Multicast

– Service model – Sample routing protocols

• QoS

– Why do we need it? – Integrated Services – Differentiated Services

• Motivated by business models